The more deeply, indeed, I consider the quinary theory (now advocated by so many talented naturalists) in all its bearings, the less consistent does it appear to me with reason and common sense; the more thoroughly am I convinced of its utter fancifulness and misleading tendency.
   Nothing in this world is without its particular and definite use, which observation, in time, generally contrives to discover. but what utility could there be, what purpose could be effected, by separate and distinct races of beings, created obviously in direct relation to particular localities, being distributed into even groups of a limited number, like the celebrated groves of Blenheim, "nodding at each other"?
   If the quinary system be universal, as some would have, pervading all creation, how is it that the stars and planets do not revolve in groups of five?  Or why even do not animals mostly produce their young by fives, or multiples of five? The absurdity is, indeed, too great to be dwelt on.  If we examine, too, the writings of even the most eminent advocates of this strange theory, we continually meet (as might be expected) with divisions apparently made for mere dividing sake, that the requisite number of groups might be filled up; and, on the other hand, with examples equally glaring of the most dissimilar forms being brought under one general head, that the same particular number should not be exceeded.
   Thus, in Mr. Selby's in many respects very valuable and useful British Ornithology, while the closely allied linnets and siskins are placed in separate subfamilies, between the types of which no supergeneric character of the least importance can be described, we find the buntings actually arranged in a subfamily of which the larks are typical; and, in another division, of like value, among his Sylviadae, four genera (Parus, Accentor, Setophaga, and Calamophilus) grouped together, which have hardly a single character in unison that is not common to the whole Dentirostres, and which, certainly , are but very distantly allied.  To adduce additional instances must be superflous: a system which can admit of such very arbitrary arrangements can have but a faint title indeed to be designated the "only natural one."

It is unnecessary now any longer to detain the attention of the reader by further prefatory observations; nor would it be worth while here to offer any remarks on the progress of plumification, the which might be better introduced as occasion may require; but I shall forthwith proceed to point out what I conceive to be of very great importance towards the classification of birds according to their true affinities, the different changes of plumage and appearance to which various groups of them are subject, confining myself, for the most part, to those upon which I can speak quite positively, from having myself had opportunities of witnessing them.
   On this inquiry there is, indeed, hardly any guide to go by, but direct personal observation; for though, in the books the greater number of these changes of appearance in the feathered race have been often mentioned, it is seldom that the precise manner in which they are brought about is stated; and the term "vernal moult" has been, in general, so very vaguely applied (sometimes indicating an actual shedding and renovation of the feathers themselves, and sometimes merely the seasonal wearing off of their winter edgings), that I have thought it best to decline altogether availing myself of their assistance.
   I may just premise, however, before commencing, that, independently of moulting, there are two principal modes by which a great alteration in the appearance of the feathers of birds is, in some cases, gradually brought about; namely, a direct change of colour in the feathers themselves, and the gradual shedding, in spring (as has already been spoken of), of their extreme tips, which are frequently of a different and more dingy colour than that part of the feather which becomes exposed to view when these have disappeared.
   A familiar and beautiful illustration of both these changes is furnished by the breast plumage of a male of the common, or song, linnet (Linaria cannabina).  The coloured portion of these feathers, in winter, is of a brownish red; and they are tipped with deciduous dusky edgings. In the spring, the latter gradually wear off, and the dark maroon changes to a bright crimson.⁴  The same plumage which the ptarmigan acquires in autumn becomes, in winter, white, and in spring gradually reassumes somewhat its former colour, but a still deeper one.⁵  Variations in general appearance, however, induced by a change of colour in the feathers themselves, are of comparatively rather unusual occurrence.⁶

There are two modes of estimating the typical standard of a natural group of species.  There are two distinct principles upon which, according as we desire to frame a system upon obvious and tangible characters, or upon the physiological relations, that is the true affinities, of species, we may arrive at very different conclusions as to which form is the more worthy to be considered the general type of the whole.
   I have said that it is not unusual to find certain characters, which, in typical forms, are comparatively little noticeable, carried out, in particular modifications of those general plans of structure, to a much greater extent; in exemplification of which may be adduced (as a familiar, though not, perhaps, the most striking instance) the great development of the bill laminae in the shoveller genus (Spathulea); also the perfection of the bill, as a groping instrument, and as a sentient organ, in the snipes and woodcocks; in consideration of which, many naturalists, esteeming these to be the most characteristic peculiarities of their respective major groups, have therefore adopted the above-named genera as the types of extensive natural families.
   Now, this may be very well in a confessedly artificial system; but, where affinity is to be considered the basis of classification, these forms will rather have to be arranged as ultimate modifications of their respective types, in a particular direction.  They are neither of them centres of radiation (at least, to any extent), such as the form of Anas boschas undoubtedly is in the duck family, and such as the godwits (Limosa) at least approximate to be in the natural family to which the snipes appertain.  Corvus and Ardea are good examples of thoroughly typical forms, which, modified in every possible way, radiate and ramify in every direction around; and so, also, is Merula, and that central division of the finch family to which the term Coccothraustes has been given.
   All of these graduate, through a series of species, into almost every form referable to their respective groups; and such must necessarily be the case with the more characteristic examples of every general plan of structure, of whatever value.  Typical forms, in fact, as a leading rule, are merely those examples of each plan which are the least bound, as a matter of necessity, to particular localities; and we accordingly find them (I mean the forms, rather than species) to be of comparatively general distribution; whereas the more one of these plans is modified to suit any particular purpose, the more completely it is adapted to any peculiar sort of locality or mode of life: the adaptation, of course, implies a receding from the general, or central, type; and the species may therefore, in technical language, be termed aberrant, even though its deviation be a farther development of characters peculiar to its group.

It is clear that we must either admit this, or allow of a multiplicity of primary types to every natural family, to every group of species framed upon the same general or leading plan: the which must necessarily lead to such gross violations of affinity as the adoption of Phasianidae and Tetraonidae of the Quinarists as separate and independent natural groups, equivalent and equally distinct from each other, as are either of them from the two contiguously ranged families, Columbidae and Struthionidae; and this, too, while the very genera assumed to be typical of them, Tetrao and Phasianus, are allied so nearly as to hybridise together.

This is so interesting a subject, that a few additional re marks may be well devoted to its elucidation. Assuming a type to be merely the abstract plan upon which a certain number of species are organised, the said plan being variously more or less modified according to the purpose for which a species was designed, it certainly does not necessarily follow that organisms simply illustrative of the mere plan should have been created, seeing that all creatures are obviously framed in direct relation to their indigenous haunts, and not as mere counterparts of one another.
   At the same time, wherever an extensive array of species are organised upon one general plan of structure, there cannot but appear some tendency to converge to a general centre; a tendency becoming more obvious as we recede from the extremes, whereupon there is usually a marked increase in the number of species exhibiting the same characters, till at length a sort of focus presents itself, as a central genus, the proper limits of which completely baffle the ingenuity of naturalists to define, inasmuch as the various species it comprises blend with, and continuously radiate into, the immediately subordinate divisions.

In illustration, it is sufficient to mention the already cited genera, Corvus, Ardea, Merula, and what should be Fringilla, but which is at present better known as Coccothraustes.  Take either of these divisions, and observe how difficult it is to define its (artificial) boundaries; how unbroken is the concatenation of species which links them with what are simply aberrant modifications of their structure, but which naturalists have been accustomed to consider as separate and distinct generic divisions.  Let us, for a moment, consider Merula.
   Some naturalists try to separate the spotted-breasted thrushes from those in which the markings are less broken; and, unquestionably, taking the extremes, there is much diversity; but there is quite as much between the different spotted-breasted thrushes.  In either case, however, where can the dividing line be drawn?  The blackbird has, when young, a spotted breast; and, in fact, the characters of its nestling plumage alone forbid its alienation from the spotted thrushes.
   Where, indeed, can we trace the line of separation between Merula and Philomela even?  And does not also the same form, in another of its various gradations, merge imperceptibly into Petrocincla, and thence into the different saxicoline genera, Erythaca, Phoenicura, and Sialia?  One ousel (Petrocincla, or, rather, Geocincla Gould) being absolutely a large robin, another a great redstart, while a newly discovered species of Sialia has the markings, and many of the characters, of a Petrocincla?
   But it would be endless to follow Merula into all its diversified ramifications. I shall content myself with tracing the series into Philomela, which is at once conclusive as to the true affinities of the latter.

To be brief, then, we observe in the European song thrush a deviation from the gregarious character of its nearest British congeners, and an approximation to the style of marking in the transatlantic species.  M. mustelina of North America is yet more solitary, and does not even associate to migrate; in this resembling Philomela, which its habits (as described by all who have observed them) accord with in almost every particular. still it retains a good deal of the true Merula; and it builds a plastered nest, like our thrushes.
   In M. solitaria the size decreases, the number of breast spots are diminished, the tarse is much lengthened (a character which commences in M. mustelina), the nest is constructed without plaster, and even the tail is rufous, as in the nightingales.  M. Wilsonii has the very form of Philomela, and is the smallest bird that ranks in Merula: its breast-spots are but very few, and these appearing as though more than half obliterated; its habits are exactly those of Philomela, and so is its nidification; and its bill hardly differs from that of our nightingale.
   The great nightingale of Eastern Europe has, according to Bechstein, an obscurely spotted breast, also a stronger bill than the common species; it is described to be more omnivorous in its diet, and, consequently, to be more hardy in a state of confinement: even its size implies an approach to the small Merulae.  And, lastly, look to the nestling plumage of the song nightingale P. luscinia), a character of no small importance in indicating the true affinities of birds, and we at once perceive its true station in the system, and how distinct it is from those forms with which (apparently from its mere size) it has been hitherto associated: it is, in fact, an ultimate modification of the type represented by Merula.

Let us now compare, for a moment, the extremes of the genus Merula; let us bring together the large mottled-backed thrushes of the East, and those diminutive solitary thrushes of the West.  Does it seem proper that these should rank in the same minimum division?  And yet how are they to be separated?  How can the former be divided from those of the missel thrush form; the last-mentioned from the fieldfare group, the fieldfares from the merles, or from the congeries to which the song thrush belongs, which last we have seen to inosculate with the nightingales?⁷  How, in like manner, can we divide the genera Ardea or Corvus?